In many experiments, experimenters are only interested in the estimation of a few factor specified effects. We call a design which can reach this target a compromise design. Partial aliased effect number pattern (P-AENP) is applicable to assess and select this kind of designs. In this article, we apply this pattern to select a catalogue of two-level compromise designs with all the factors divided into two groups, main effects of one group and two-factor interactions between groups are concerned. Some theoretical results are obtained, and a number of clear and general optimal compromise designs with 8, 16, 32 and 64 runs are tabulated.

Zhongxu Zhang

Papers

Research on Compromise Designs of Two-Group Factors with Consideration of Certain Main Effects and Two-Factor Interactions Between Groups